Project Summary The overall goal of this application is to develop hypoxia-derived prognostic biomarkers to predict breast cancer progression and patient outcome from human breast cancer tissue samples. In a previous NIH-funded project, have identified hypoxia-driven signaling networks in the breast tumor microenvironment (TME) through multimodal imaging combined with omics approaches in preclinical models. The breast TME contains several spatially heterogeneous hypoxic regions, which induce metastasis and drive angiogenesis, invasion, altered metabolism, and resistance to radiation and chemotherapy. Hypoxic tumor regions are also known to select for aggressive cancer phenotypes with the highest capacity for metastatic spread. We have identified significantly changed molecular signatures in hypoxic tumor regions in human breast tumor xenograft models, with confirmed lists and networks of metabolites, lipids, and proteins that are significantly altered. We propose to now build on these studies by developing fast reproducible clinical sample preparation and mass spectrometry imaging (MSI) protocols that allow for high throughput use in clinical pathology laboratories and seamlessly integrate with histology and immunohistochemistry. We will test in large cohorts of human breast cancer tissue samples if hypoxic metabolome, lipidome, and proteome signatures have prognostic clinical value. To this end, we will employ innovative multi-enzyme on-tissue digestion for proteins and glycans, reactive desorption electrospray ionization (DESI) for enhanced metabolic marker discovery, and MSI-based Ozonolysis (OzID) for on-the-fly lipid isomer imaging. In Aim 2, we will use these MSI approaches for analyzing tissue microarrays with biopsies from the primary tumors of over 1,000 breast cancer patients to test if hypoxic molecular signatures can predict patient outcome, recurrence, and 5-year-survival. In Aim 3, we will address the most life-threatening aspect of breast cancer, the formation of metastases, and investigate if hypoxic regions in primary human breast tumors are driving the development of metastases. The proposed research will identify key molecular networks through which hypoxia drives breast cancers towards worse outcome and metastasis. Recent developments in MSI have significantly improved its imaging speed, making it now possible to perform MSI-based molecular pathology in clinically relevant times. This enables us to translate the results of our earlier studies on the effect of hypoxia in the breast TME directly to large patient cohorts. We will clinically evaluate matrix-assisted laser desorption/ionization (MALDI) MSI as diagnostic tool in conjunction with the routine clinical pathology workup for accurate molecular prognosis of breast cancers. The large-scale metabolite and lipid signatures obtained in our application will support the emerging use of ambient ionization MSI applications for accurate intraoperative margin detection during breast-conserving surgery and intraoperative diagnostics using the iKnife.